View Full Version : Controlling castback pump speed with temperature?

02-16-2015, 12:55 PM
Has anyone tried using a temperature sensor and VFD to control the castback pump speed--and hence the castback temperature? Just something that came up over a few post-shift beers with the brewers the other night.

02-17-2015, 01:14 PM
I used to work in a regional craft brewery- the larger brewhouse had VFDs on both the wort pump and cold liquor pumps that would respond to both temperature and flow-rate setpoints. It was definitely a nice feature, mostly worked well and also allowed for precise manual control when necessary (automation is great when it works properly!).
It was a computer controlled brewhouse, but I imagine it could be accomplished with simpler PLC controls?

02-17-2015, 02:36 PM
Our VFD has a 4-20 mA control input. I'm looking for a temp indicator to run off an RTD probe that has a 4-20 mA output. Then figure out how to configure the critter....

02-18-2015, 08:07 AM

Yes, that is what we have. The cold liquor pump runs wide open, and the VFD on the wort pump is controlled by the temperature of the exiting wort. I don't know what kind of temperature controller you have, but I would guess it would need to be able to run in PID mode, with a 4-20 mA or 0-10VDC output to the VFD.

02-18-2015, 09:17 AM
Thanks, Linus. I'm working with a couple of folks on putting something like that together. Any chance you could get a name or nuber off your temp controller? I have a couple of spare Fujis sitting around I might be able to use.

02-18-2015, 10:04 AM

Ours is part of a larger computer system, not a standalone. It's been a while, in another life, but I've set up a small PID controller to output to a VFD before. I don't remember it being too difficult. The trick is finding the right set of P, I, and D to control the process if it doesn't autotune itself.

02-19-2015, 11:10 AM
The digital VFDs we use on all of our portable pumps have an optional IODA input card (about $125) which allows them to be controlled from a remote 4-20mA signal like a temperature or pressure sensor. It is a pretty simple set up and you can switch back to manual mode at the touch of a button for other uses.

Here is a link to the pages on our website

22238 (http://www.cpesystems.com/collections/vfds)

Digital VFD's (http://www.cpesystems.com/collections/vfds)

22237 (http://www.cpesystems.com/collections/vfds/products/input-output-multi-function-board)

IODA Board (http://www.cpesystems.com/collections/vfds/products/input-output-multi-function-board)

Please feel free to call us if you have any questions.

02-19-2015, 11:40 AM
Has anyone tried using a temperature sensor and VFD to control the castback pump speed--and hence the castback temperature? Just something that came up over a few post-shift beers with the brewers the other night.

Timm, what exactly is a " castback " pump?
No one one this side of the Rio Grande has ever heard that one before.

02-19-2015, 01:11 PM
CPE folks--We just replaced our VFD for castback and CIP. I think the unit we purchased has the IO board built in.

Warren--aren't we on the same side of that minor waterway?

I'm not sure what "castback" would be called elsewhere, but it's what we provincials call the transfer from the whirlpool through the HX to the ferm. We pump push from the whirlpool to the HX, and the speed of the pump controls the output temp of the wort from the HX. Seems like a no-brainer to automate this, but right now, it's starting to seem too expensive to do it the way I'd like.

Any thoughts appreciated....

02-19-2015, 02:06 PM
Lots of folks (me, included) refer to it as "knockout." Tomay-to, Tomah-to....

I previously worked at a large regional with a 120bbl Huppmann brewhouse. The system was designed for the knockout pump to run at full speed and temperature was controlled by a PLC unit that controlled a restricting valve on the chilling water. That way the speed of delivery of the wort remained constant and the oxygenation (through a venturi system) wasn't affected.

Might be easier trying to do it that way?

Cheers- Mike

02-19-2015, 07:55 PM
Timm, trying to figure out what problem you have that you're trying to solve. If you are just trading one control parameter for another, I'd argue that length of time in the kettle at high temperature is more important than the exact knockout temperature. The knockout temperature could be plus or minus a few degrees without any real change in the beer profile. I wouldn't say that with respect to length of time in the whirlpool at temperature (DMS, hop profile, etc.). Can't you take the last bit of heat from the wort with the fermenter jackets? I'm all in favor of keeping it simple. Best of luck!

02-20-2015, 10:22 AM
Phillip--We use mains water to cool our HX (which we recover to feed our HL system). Because the flow rate varies depending on other water demands in the brewery, keeping the output temp of the wort within the desired range for pitching requires near constant monitoring of the temperature and adjustment of the pump speed.

I've got a couple of quotes for an electronic thermometer with 4-20mA output to control the freq drive, but they are a bit on the high side. I'm doing a little research to see if I can put together a system for little less, but may just have to bite the bullet.

02-20-2015, 04:44 PM
We have the same issues in Palau. So we just don't use the water during the 30 minutes of knockout. Not a big deal. But we are planning for a CLT and pump that should eliminate this minor issue. I don't see any issues with your idea, however. Why 4-20mA input? Most any cheap controller will have inputs for many devices; including RTD or thermocouple. And output for 4-20mA devices like a VFD. I've done this with glycol pump to control pressure before. Your application should be relatively easy. Let me know what you have and I'm relatively sure I can make it work. Best of luck!

02-22-2015, 12:43 PM
Thanks, Phillip.

I'm really surprised by the size of estimates I've gotten to build a controller for this, especially since I've realized that pretty much any temperature controller will give me the 4-20mA output I need to control the VFD. I've got two quotes, one for ~$700, the other for over $1,000. All I need is a platinum 3-wire RTD probe, a NEMA 4 enclosure, and a simple temperature controller--well less than $300. For this price, this set-up is competitive with high-accuracy digital thermometers, not to mention controllers.

Unfortunately, the Fuji PXR controllers we use for our ferms do not have the 4-20mA output, since I keep a couple of spares on hand.

You'll probably be hearing from me when it comes to interfacing the control with the VFD--hope you don't regret your offer!

02-22-2015, 05:47 PM
Literature I have for PXR series (PXR3-9) shows 4-20 output as available on all models for both heating and cooling modes. Perhaps your particular model doesn't include it, but that seems odd. A new PXR can be had for less than $100 on ebay. And another $20 for a pT100 probe. Seems like those two components, some wire, and an afternoon would wrap this up. The hard part would be programming for a delayed feedback, but autotune should be able to do that. Best of luck!

02-23-2015, 05:19 AM
I am now educated

I used to be from the Texas side.

Best of luck on the project

02-23-2015, 04:58 PM
Is this variability due to flow rate or changes in pressure? In my case the pressure would change, thus changing the flow rate. I added a pressure regulator in my main, before the heat exchanger, set it lower than the lowest I ever see our water pressure. This really smoothed out my ups and downs.

Rich DeLano

02-24-2015, 08:23 AM
Thanks, Warren. I've also been educated--knockout if that's the word everyone else uses.

Rich--excellent idea! I'll stick a pressure gauge on the cold water line next time we brew.

This idea grew from my effort to find a truly accurate digital thermometer for the knockout (and the mash tun). We're all tired of having to calibrate the bimetal therms every brew, and it seems it's getting harder to find decent quality bimetal therms. After doing too much research, it seems the best way to get the reliability and accuracy we want is to use platinum RTDs and good quality digital temperature indicators/controllers. Since I'm already forking out for the controller, I might as well use it to control something, and the brewers thought this was a great idea.

I'm still surprised no one makes a proper digital brewery thermometer. The only ones I've found are homebrew jobs, which actually are nothing but digital food thermometers with an accuracy, generally, of ~+-2* F, and who knows what actual precision. Then there are the FDA approved process therms for the dairy and meat processing industries, with all the legal logging and approvals and the consequent costs.